Back to Details

agile-onboarding

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Onboarding

入职引导

Use this skill to guide new team members through the agile + AI flow, in a practical and progressive way.
使用本技能以实用且循序渐进的方式引导新成员了解敏捷+AI工作流。

Objective

目标

  • Provide context about the operational model (Light Scrum + AI as pair)
  • Teach the artifact flow in practice, not theory
  • Ensure the new member can operate autonomously in 1-2 sprints
  • Avoid onboarding being just documentation — it must include practice
  • 介绍“轻量Scrum + AI协作伙伴”的运营模式背景
  • 教授实际工作中的工件流转，而非理论知识
  • 确保新成员能在1-2个sprint内独立开展工作
  • 避免入职仅停留在文档层面——必须包含实践环节

When to use

使用场景

  • New dev or manager joins the team
  • Someone changes roles (dev becomes tech lead, for example)
  • The team adopts the flow for the first time
  • Someone needs retraining after time away
  • 新开发人员或经理加入团队时
  • 成员角色变更时（例如开发人员转为技术负责人）
  • 团队首次采用该工作流时
  • 成员离岗后需要重新培训时

Onboarding trail

入职流程

Day 1: Understand the model

第1天：理解流程模型

Objective: know what exists and why.
  1. Present the complete flow:
    mermaid
    flowchart LR
    A["/agile-intake"] --> B["/agile-roadmap"]
    B --> C["/agile-epic"]
    C --> D["/agile-story"]
    D --> E[execution]
    E --> F["/agile-status"]
    F --> G["/agile-retro"]
  2. Explain the role division:
    • Human: decides, validates, controls git, communicates
    • AI: structures, implements, verifies, reports
  3. Show the decision tree:
    • When to use task vs epic
    • Scope assessment (small to large)
  4. Show the available skills and how to invoke each one:
    • /agile-intake
      — capture problems
    • /agile-roadmap
      — strategic direction
    • /agile-epic
      — decompose and structure initiatives
    • /agile-story
      — execution plans
    • /agile-refinement
      — validate artifacts and review code
    • /agile-status
      — track progress (checkpoint, consolidation, closure)
    • /agile-sprint
      — sprint planning
    • /agile-review
      — sprint review and demo
    • /agile-metrics
      — sprint metrics
    • /agile-retro
      — retrospective
    • /agile-proto
      — interactive prototypes
    • /agile-router
      — guidance on which skill to use
目标： 了解现有流程及设计逻辑。
  1. 展示完整工作流：
    mermaid
    flowchart LR
    A["/agile-intake"] --> B["/agile-roadmap"]
    B --> C["/agile-epic"]
    C --> D["/agile-story"]
    D --> E[execution]
    E --> F["/agile-status"]
    F --> G["/agile-retro"]
  2. 解释角色分工：
    • 人类：决策、验证、管理git、沟通协调
    • AI：结构化梳理、执行实现、验证检查、生成报告
  3. 展示决策树：
    • 何时使用task（任务）vs epic（史诗）
    • 范围评估（从小到大）
  4. 展示可用技能及调用方式：
    • /agile-intake
      — 问题收集
    • /agile-roadmap
      — 战略方向规划
    • /agile-epic
      — 拆解并结构化举措
    • /agile-story
      — 执行计划制定
    • /agile-refinement
      — 工件验证与代码评审
    • /agile-status
      — 进度跟踪（检查点、汇总、收尾）
    • /agile-sprint
      — 迭代规划
    • /agile-review
      — 迭代评审与演示
    • /agile-metrics
      — 迭代指标分析
    • /agile-retro
      — 回顾总结
    • /agile-proto
      — 交互式原型制作
    • /agile-router
      — 技能使用指引

Day 2: Practical exercise — intake and planning

第2天：实践练习——需求收集与规划

Objective: create an intake and a simple plan with AI support.
Suggested exercise:
  1. The new member chooses a small, real problem (bug, improvement, task)
  2. Uses the 
    /agile-intake
    skill to structure the problem
  3. Decides the correct artifact with the decision tree
  4. Uses 
    /agile-router
    to validate the choice, then creates the plan with 
    /agile-story
    or 
    /agile-epic
  5. The mentor/tech lead reviews and gives feedback
目标： 在AI支持下完成需求收集与简单计划制定。
建议练习：
  1. 新成员选择一个小型实际问题（bug修复、功能优化、日常任务）
  2. 使用
    /agile-intake
    技能梳理问题
  3. 通过决策树选择合适的工件
  4. 使用
    /agile-router
    验证选择，随后通过
    /agile-story
    或
    /agile-epic
    创建计划
  5. 导师/技术负责人进行评审并给出反馈

Day 3: Practical exercise — execution with TDD

第3天：实践练习——结合TDD的执行环节

Objective: implement something using the task -> TDD -> verification flow.
Suggested exercise:
  1. Take the plan created on day 2
  2. Implement using TDD with AI as pair:
    • Describe the expected behavior
    • AI writes the test (red)
    • AI implements (green)
    • Dev requests refactoring if necessary
  3. Run verifications (lint, types, tests)
  4. Run 
    /agile-refinement
    (code review mode) to review the diff before committing
目标： 遵循“任务 -> TDD -> 验证”流程完成功能实现。
建议练习：
  1. 基于第2天创建的计划
  2. 与AI协作采用TDD方式实现：
    • 描述预期行为
    • AI编写测试用例（红阶段）
    • AI完成功能实现（绿阶段）
    • 开发人员按需发起重构请求
  3. 执行验证（代码检查、类型校验、测试）
  4. 在提交前使用
    /agile-refinement
    （代码评审模式）审查代码差异

Day 4: Practical exercise — tracking

第4天：实践练习——进度跟踪

Objective: generate status updates and close with a report.
  1. Use 
    /agile-status
    (checkpoint mode) to generate a progress update
  2. Simulate a 
    /agile-status
    (consolidation mode) report for the period
  3. Close the delivery with 
    /agile-status
    (closure mode)
  4. Review the complete chain: task -> execution -> status -> closure
目标： 生成状态更新并完成收尾报告。
  1. 使用
    /agile-status
    （检查点模式）生成进度更新
  2. 模拟
    /agile-status
    （汇总模式）生成周期报告
  3. 使用
    /agile-status
    （收尾模式）完成交付闭环
  4. 回顾完整链路：任务 -> 执行 -> 状态 -> 收尾

Day 5: Reflection and autonomy

第5天：复盘与独立能力评估

Objective: assess if the new member is ready to operate autonomously.
  1. The new member conducts an intake alone
  2. Creates plan or epic without mentor help
  3. Implements with TDD
  4. Closes with status report
  5. Mentor validates and gives final feedback
目标： 评估新成员是否具备独立工作能力。
  1. 新成员独立完成需求收集
  2. 在无导师协助的情况下创建计划或epic
  3. 采用TDD方式完成实现
  4. 生成状态报告并完成收尾
  5. 导师进行验证并给出最终反馈

Onboarding checklist

入职检查清单

  • Understands the complete flow (intake to retro)
  • Knows how to choose the right artifact (decision tree)
  • Can create task or epic with AI support
  • Knows how to use TDD with AI as pair
  • Knows how to generate status updates and closure reports
  • Understands the responsibility division (human vs AI)
  • Knows which skills are available and when to use each one
  • Completed at least one full cycle (intake -> closure) with supervision
  • 理解完整工作流（从需求收集到回顾总结）
  • 掌握如何通过决策树选择合适的工件
  • 能在AI支持下创建task或epic
  • 掌握与AI协作的TDD方法
  • 能生成状态更新及收尾报告
  • 理解人机角色分工
  • 了解可用技能及适用场景
  • 在监督下完成至少一次完整流程（需求收集 -> 收尾）

Adaptation by profile

按角色适配内容

For devs

开发人员

  • Focus on: TDD, pair programming with AI, quality gates, git workflow
  • Extra exercise: implement a small feature from scratch using the flow
  • 重点关注：TDD、与AI结对编程、质量门禁、git工作流
  • 额外练习：使用该工作流从零实现一个小型功能

For managers / scrum masters

经理/Scrum Master

  • Focus on: roadmap, epic decomposition, sprint planning, retro, status reports
  • Extra exercise: conduct an epic decomposition and sprint planning with AI support
  • 重点关注：路线图规划、epic拆解、迭代规划、回顾总结、状态报告
  • 额外练习：在AI支持下完成epic拆解与迭代规划

For tech leads

技术负责人

  • Both focuses: planning and execution
  • Extra exercise: review AI-generated code with 
    /agile-refinement
    and give constructive feedback
  • 兼顾两类重点：规划与执行
  • 额外练习：使用
    /agile-refinement
    评审AI生成的代码并给出建设性反馈

Rules

规则

  • Onboarding is not passive. The new member must practice, not just read.
  • The mentor does not do it for the new member — guides and reviews.
  • Onboarding mistakes are opportunities, not failures. The environment must be safe to experiment.
  • If the new member cannot complete the checklist in 5 working days, the problem may be the process, not the person. Discuss in retro.
  • 入职不是被动学习，新成员必须动手实践，而非仅阅读文档
  • 导师不能代劳，仅提供指导与评审
  • 入职中的错误是成长机会，而非失败，需营造安全的实验环境
  • 如果新成员无法在5个工作日内完成检查清单，问题可能出在流程而非个人，需在回顾会议中讨论

Relationship with the flow

与工作流的关联

This skill acts as the entry point to all others. After onboarding, the member should be able to invoke 
/agile-router
, 
/agile-story
, 
/agile-status
, 
/agile-sprint
, and 
/agile-retro
autonomously.
本技能是所有其他技能的入口。完成入职后，成员应能自主调用
/agile-router
、
/agile-story
、
/agile-status
、
/agile-sprint
和
/agile-retro
。